Before implementation of disease prevention programs, the incidence of neonatal early-onset group B streptococcal (GBS) sepsis was 1.7 per 1,000 births with a fatality rate approaching 50%.1,2 Although maternal colonization rates differ according to ethnicity, approximately 20% of pregnant women are colonized with GBS, placing their newborns at risk for acquiring infection during delivery.3 Past studies have demonstrated that the administration of antimicrobial prophylaxis during labor to women at risk for transmitting GBS to their neonates has reduced the rate of early-onset GBS disease by 50%.4–7
The majority of studies used intrapartum prophylaxis in women who were colonized in the third trimester as determined by anogenital testing and presented with risk factors for transmission.8,9 Other studies have shown that postnatal prophylaxis alone administered to newborns provided some protective effect as well.10,11 The largest experience with postnatal prophylaxis demonstrated a 60% reduction in early-onset GBS disease.12,13
Current Centers for Disease Control and Prevention,14 American Academy of Pediatrics,15 and American College of Obstetrics and Gynecology16 guidelines recommend universal maternal screening at 35–37 weeks of gestation with selective intrapartum prophylaxis to colonized women.14 A risk factor approach, previously endorsed as equally acceptable by all three administrative bodies, involved the administration of chemoprophylaxis to all women in labor with rupture of the membranes greater than 18 hours, intrapartum fever greater than 38.0°C, and anticipated preterm delivery (less than 37 weeks of gestation). In both approaches, antimicrobials are given to women with antepartum GBS bacteriuria or a prior neonate with known GBS disease. Given the paucity of data concerning the efficacy of any strategy, Schrag et al17 performed a retrospective cohort study and estimated the maternal universal screening strategy was 50% more effective than the risk-based protocol. Unfortunately, recent studies have demonstrated failures in the culture-based strategy. Two recent studies found that most early-onset GBS sepsis cases now are in term neonates whose mothers screened negative at 35–37 weeks of gestation.2,18 Are these false-negative cultures or the product of postculture colonization? Recent studies have also reported the negative predictive value of the current GBS culture-based strategy in predicting intrapartum GBS colonization status to be 88–95% with a 10% false-negative rate.19,20 These findings give credence to continued investigation of a nonculture-based strategy for GBS prevention.
In 1995, a combined maternal and neonatal protocol for the prevention of early-onset GBS was developed at Parkland Memorial Hospital. At this time, there was a lack of consensus about the efficacy of GBS cultures in late pregnancy and disease prevention. The challenges of implementing a screening-based strategy at Parkland Memorial Hospital included the need to process over 15,000 cultures yearly with timely results, many of which would be collected from various off-site prenatal care centers. A risk-based strategy alone would fail to provide prophylaxis to the full-term neonate born from a colonized mother without identifiable risk factors, important in a setting where the majority of births are low risk and term. The protocol incorporated a risk based approach with the addition of prophylactic intramuscular penicillin (intramuscularly) given to newborns within 1 hour of delivery. After implementation of the combined obstetric and neonatal chemoprophylaxis protocol, the rate of early-onset GBS decreased from 1.95 cases per 1,000 live births during the years 1986–1994 when no prophylaxis was used to .47 cases per 1,000 live births during the years 1995–1999 in over 68,000 deliveries. There was no increase in disease caused by other organisms.21
The purpose of this study is to estimate the rates of GBS disease using this combined protocol with focus on compliance and assessment of continued efficacy at Parkland Memorial Hospital between the years 2000 and 2008. In addition, rates of β-lactam antibiotic resistance in Gram-negative infections were estimated during these years.
MATERIALS AND METHODS
The GBS chemoprophylaxis protocol was implemented in January 1995 and continued throughout the study period between January 1, 2000, and December 31, 2008. The protocol consisted of administration of intrapartum intravenous ampicillin (2 g every 6 hours) to all mothers with any of the following risk factors: 1) preterm labor or anticipated delivery before 37 weeks of gestation; 2) rupture of membranes for more than 18 hours; 3) intrapartum fever of 38°C or greater (chorioamnionitis); 4) previous delivery of a neonate with GBS infection; and 5) history of GBS bacteriuria during the current pregnancy. In addition to ampicillin, women with chorioamnionitis received gentamicin (loading dose of 120 mg followed by 80 mg intravenously every 8 hours). Clindamycin was provided to women who reported allergy to β-lactam antibiotics. The neonatal component consisted of a single intramuscular dose of aqueous penicillin G administered in the delivery room within 1 hour of birth to all newborns with no clinical signs or symptoms of infection. The dose of penicillin administered was 60,000 units (1 mL) for full-term newborns and 30,000 units (0.5 mL) if the birth weight was less than 2 kg. Neonates with clinical signs or symptoms of infection and those born to women who received ampicillin and gentamicin for chorioamnionitis did not receive penicillin prophylaxis in the delivery room but were treated with ampicillin and gentamicin after a blood culture was obtained.21,22
Women with preterm premature rupture of membranes received ampicillin and gentamicin for the first 48 hours of admission. Labor was induced at 34 weeks of gestation in this situation unless it was indicated or spontaneously occurred before this time. During labor, these women received ampicillin for GBS prophylaxis.
All neonates who were born at Parkland Memorial Hospital from January 1, 2000, to December 31, 2008, and who developed early-onset (occurring at 72 hours of age or younger) invasive bacterial disease caused by penicillin-susceptible and -resistant pathogens were identified using the continuously updated Parkland Memorial Hospital obstetric operations database used for research purposes23 supplemented by disease-specific International Classification of Diseases, 9th Revision codes. Corresponding maternal data were also identified in this manner. These neonates and their mothers were also prospectively identified by the Pediatric Infection Control Group under institutional review board approval from the University of Texas, Southwestern. Additional clinical maternal and neonatal information was abstracted from chart review.
Early-onset GBS infection was defined by the isolation of GBS from blood or cerebrospinal fluid from a newborn 72 hours of age or younger. Three days was used to define early-onset disease because greater than 95% of cases of GBS infection that occur at this institution within the first week of life occur at 72 hours of age or less.12,24,25
Incidence rates for sepsis resulting from GBS along with other organisms were estimated. Compliance with risk factor identification and appropriate treatment were also estimated. Rates of β-lactam resistance among neonatal disease caused by Gram-negative organisms was calculated.
Rate and exact 95% confidence intervals are presented using standard methods for rates and proportions. Pearson χ2 was used to compare rates. P values <.05 were considered significant.
There were 94 cases of early-onset GBS sepsis among 143,467 live births at Parkland Memorial Hospital between January 1, 2000, and December 31, 2008 (0.66 per 1,000 live births [0.53–0.80 per 1,000]). Cumulatively, the overall rate since the new protocol was introduced (1995–2008) was 0.59 (0.49–0.71 per 1,000). Cases by year with current protocol are displayed in Figure 1 in gray (black denotes years 1972–1994). Demographic and clinical characteristics of these newborns and mothers are shown in Table 1. Of the cases that underwent cesarean delivery, one cesarean delivery was performed before rupture of membranes.
Eighty-two (87%) of the 94 neonates with early-onset GBS were born to women at term (more than 37 weeks of gestation) and 47% of the neonates did not display clinical signs or symptoms of infection. The most common presenting symptom was respiratory distress (n=42 [44%]). Those without signs or symptoms were ascertained during neonate workups for maternal chorioamnionitis. One neonate was diagnosed with bacteremia or sepsis by cultures drawn after birth; and the remaining neonate was diagnosed on day 3 of life after the development of respiratory distress.
The distribution of maternal risk factors is listed in Table 2. Intrapartum fever was the most common maternal risk factor (n=63 [67.0%]) followed by prolonged rupture of the membranes (n=28 [29.7%]). Although most mothers had one or more risk factors for GBS transmission, 20 (21.2%) did not have risk factors. Of the 74 women with maternal risk factors, 62 (84.0%) received intrapartum chemoprophylaxis; 12 women did not receive antibiotics, prophylactic or therapeutic (five chorioamnionitis, five prolonged rupture of membranes, and two preterm neonates). Most women with chorioamnionitis (n=58 [91%]) who delivered newborns that developed early-onset GBS sepsis received intrapartum prophylaxis. Of the five that did not receive prophylaxis, four were diagnosed with chorioamnionitis within minutes of delivery before antibiotic therapy could have been initiated. Time of fever was not documented in the remaining case. In the five women with prolonged rupture of the membranes who did not receive chemoprophylaxis, four were precipitous labor and deliveries that arrived to the labor and delivery unit moments before delivery. One woman was not initiated on antibiotics and delivered after 20 hours of ruptured membranes. Only two women with preterm birth did not receive chemoprophylaxis. The first was a result of a precipitous labor and delivery, and the second patient did not receive antibiotics secondary to erroneous gestational age assessment ascertained postpartum.
Ninety-five percent of the neonates with early-onset GBS disease without risk factors were symptomatic at birth; therefore, penicillin prophylaxis was held and intravenous antibiotics were started. Antimicrobial susceptibility for GBS isolates was ordered at the discretion of the admitting neonatologist or pediatrician based on the clinical scenario; and, consequently, results are available for 45 of the 94 samples. Four (8.8%) of the GBS isolates demonstrated resistance to clindamycin and 12 (26.6%) were resistant to erythromycin. Five neonates demonstrated evidence of bacteremia from other organisms concurrent with GBS. Two had Escherichia coli sensitive to β-lactam antibiotics, two were had Enterococcus infection without available susceptibility reports, and one had Haemophilus parainfluenzae, also without susceptibility reporting available. During this study period (2000–2008), there was one death from early-onset GBS infection, no cases of meningitis, and the median length of stay for these neonates was 8 days. The one death occurred in a 610-g neonates born at 25 weeks of gestation after 4 weeks of ruptured membranes.
There were 34 cases of early-onset sepsis from other organisms during the study period for a rate of 0.24 per 1,000 live births, respectively. Rates per year are demonstrated in Figure 2. Only five of the women with neonates that developed early sepsis from other organisms did not have risk factors (15%). Twenty-three (67%) culture results with susceptibility reporting were available from the 34 neonates with non-GBS early-onset sepsis. Of the available culture results with reported susceptibilities, 15 (65.2%) were sensitive to ampicillin. Of the 19 neonates with early-onset Gram-negative sepsis and available susceptibility results, eight (42.1%) were sensitive to β-lactams.
This report updates a single institution's 30-year experience investigating the prevention of neonatal GBS sepsis. Experience in the past 9 years has shown a sustained reduction in GBS infections in newborns using a combined maternal and neonatal prevention strategy. Importantly, with continued use of this protocol, there was no increase in sepsis caused by other organisms.10–12
Between the years 1986 and 1994, rates of early-onset GBS peaked at 2.2 per 1,000 live births. In 1995, the implementation of the combined maternal and neonatal chemoprophylaxis protocol resulted in an over 80% reduction in neonatal GBS infection with rates of early-onset disease as low as 0.40 per 1,000 births.21 In addition to the dramatic decrease in GBS disease, mortality from this infection decreased from 10% to 0% with no increase in disease caused by other pathogens. Continued surveillance over the next 4 years revealed similar rates of early-onset GBS disease (0.47 per 1,000 births) with no neonatal deaths using the combined maternal and neonatal chemoprophylaxis strategy.22
With adherence to current Centers for Disease Control and Prevention-based guidelines concerning GBS, the overall rate of early-onset GBS has been 0.34–0.37 per 1,000 live births over the past several years.14 However, rates from 2003 to 2006 among African American neonates increased significantly from 0.53 to 0.86 cases per 1,000 live births.26 This failure may be explained by the challenges surrounding effective methods to ensure implementation of the culture based protocol, especially in women who are higher risk with inadequate prenatal care. In a recent study, the Centers for Disease Control and Prevention sponsored a multistate retrospective analysis of national compliance to the culture-based method. Antenatal screening and the appropriate administration of chemoprophylaxis nearly doubled to 85% from 1999 to 2004; however, the largest portion (74.4%) of cases of early-onset GBS occurred in term neonates, of which 61.4% of their mothers screened negative during routine antenatal screening.2 In another recent review, 82% of mothers of term neonates that developed invasive disease screened negative during their antenatal care. Many of these mothers did not demonstrate clinical risk factors for GBS colonization or infection.18
In this 9-year study cohort, we have found the rate of early-onset GBS disease to be 0.66 per 1,000 births. Although this is slightly increased from the proceeding 5 years, there is no significant difference in rates between the years 1995 and 2008 (0.47 per 1,000 compared with 0.66 per 1,000, P=.09; Fig. 1) in over 200,000 deliveries. The current case fatality rate is approximately 5–7%,26 so it is encouraging that there has been only one death associated with early-onset GBS since the implementation of our protocol in 1995. The neonatal death occurred in a 610-g neonate born at 25 weeks of gestation after 4 weeks of ruptured membranes. There were no cases of neonatal meningitis in the early-onset GBS group. This lack of major morbidity and mortality suggests an advantage to our dynamic process of a combined maternal and neonatal prophylaxis strategy compared with a static screening protocol.
A concern about the safety of combined maternal and neonatal GBS prophylaxis regimens has been that the universal administration of antibiotics for each birth may selectively increase non-GBS sepsis or increase antimicrobial-resistant sepsis. The rate of sepsis caused by other pathogens did not increase over the study period. Furthermore, compared with the previous 5 years, rates of early sepsis caused by non-GBS pathogens were slightly lower at a rate of 0.24 compared with 0.62 per 1,000 births (P<.001). Of the 34 non-GBS sepsis neonates in this study, 23 had Gram-negative sepsis (0.16 per 1,000 births). This rate is also lower than the rate of Gram-negative sepsis reported between the years 1995 and 2000 (0.30 per 1,000 live births, P=.02). These are lower rates than reported elsewhere.26–28 Potential biases in our report include underascertainment of all non-GBS early-onset sepsis and there may be a certain incidence of culture-negative sepsis (a condition more likely with neonatal prophylaxis12).
Of Gram-negative isolates with reported susceptibilities, 50% were sensitive to β-lactam antibimicrobials. The non-GBS isolates that were not Gram-negative included: coagulase-negative Staphylococcus species, Enterococcus faecalis, methicillin-sensitive Staphylococcus aureus, Listeria monocytogenes, and Candida species. There was only one non–GBS-associated death in a 770-g newborn with Serratia marcescens neonatal sepsis who died at day of life 110 with Candida sepsis. Overall, during the past 9 years, the death rate from early-onset sepsis has been 0.014 per 1,000, a 21-fold decrease since that reported by Siegel and Cushion for the years 1986 through 1994 at Parkland Memorial Hospital. Our incidence of ampicillin-resistant non-GBS sepsis has not increased, and this is consistent with findings from Puopolo et al.26
Other concerns regarding universal intramuscular penicillin G to all newborns revolve around neonate safety and cost. We have previously demonstrated the efficacy and safety of routine penicillin G administration to term neonates.12 In addition, unlike adults, neonates do not develop penicillin-induced anaphylaxis.12 Moreover, we have noted a consistent parental acceptance of a single intramuscularly injection at the time of delivery. The current cost to Parkland Memorial Hospital for a single dose of 60,000 units of penicillin G is 16 cents. To provide prophylaxis to 15,000 births a year would cost $2,400. This is clearly less than the cost of treating one neonate with early-onset GBS sepsis. Prevention of a single case of early-onset GBS disease would save more than the total cost of prophylaxis of 15,000 neonates.
At the time when both the risk-based approach and the screening-based approach were equally accepted for GBS prophylaxis, it was evident that using the culture-based method for GBS prophylaxis would pose significant obstacles and adversely affect patient care at our institution. Mainly, handling specimens and communicating culture results from over eight off-site centers in a timely fashion would be difficult. In addition, the cost of processing over 15,000 cultures when the efficacy of the screening-based protocol had not been firmly established was a significant deterrent to implementing this strategy. An additional advantage of the combined maternal and neonatal protocol is that it provides prophylaxis to the full-term neonate born to a mother without risk factors whose carrier status is unknown. This is an important consideration given that over 40–70% of neonates with early-onset GBS are born to low-risk women at term.2,18 With recent data illustrating GBS disease occurrence in neonates of women with negative screening, a strategy including postnatal prophylaxis should be strongly considered for disease prevention.2
Institutions that face similar challenges of specimen handling, result reporting, and cost of a screening-based approach should consider this strategy of maternal and neonatal chemoprophylaxis for the prevention of early-onset GBS disease. Although these results are encouraging, they are only possible, like with any prevention program, with health care provider compliance.
During the 9-year study period, there were only four violations of the protocol in neonates who contracted early-onset GBS sepsis, most of which resulted from inaccurate antepartum or intrapartum gestational age assessment. Ultimately, success of any prevention strategy depends on compliance. Although this uncontrolled study does not definitively prove the efficacy of our strategy, the sustained rates in early-onset GBS sepsis along with the low rates of neonatal disease caused by other pathogens over the past 13 years demonstrate the feasibility and durable efficacy of our combined maternal and neonatal GBS chemoprophylaxis protocol for the prevention of GBS disease.
1. Franciosi RA, Knostman JD, Zimmerman RA. Group B streptococcal neonatal and infant infections. J Pediatr 1973;82:707–18.
2. Van Dyke MK, Phares CR, Lynfield R, Thomas AR, Arnold KE, Craig AS, et al.. Evaluation of universal antenatal screening for group B streptococcus. N Engl J Med 2009;360:2626–36.
3. Siegel JD. Prophylaxis for neonatal group B streptococcal infections. Semin Perinatol 1988;22:33–49.
4. Schrag SJ, Zywicki S, Farley MM, Reingold AL, Harrison LH, Lefkowitz LB, et al.. Group B Streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 2000;342:15–20.
5. Boyer KM, Gotoff SP. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum prophylaxis. N Engl J Med 1986;314:1665–9.
6. Morales WJ, Lim D. Reduction in group B streptococcal and neonatal infections in preterm pregnancies with premature rupture of the membranes through a rapid identification test. Am J Obstet Gynecol 1987;157:13–6.
7. Garland SM, Fliegner JR. Group B streptococcus and neonatal infections. The case for intrapartum prophylaxis. Aust N Z J Obstet Gynecol 1991;31:119–22.
8. Pylipow M, Gaddis M, Kinney JS. Selective intrapartum prophylaxis for group B streptococcus colonization: management and outcomes of newborns. Pediatrics 1994;93:631–5.
9. Gibbs RS, McDuffie RS, McNabb F, Fryer GE, Miyoshi T, Merenstein G. Neonatal group B streptococcus sepsis during 2 years of universal screening in pregnancy. Obstet Gynecol 1994;84:496–500.
10. Siegel JD, McCracken GH Jr, Threlkeld N, Milvenan B, Rosenfeld CR. Single dose penicillin prophylaxis of neonatal group B streptococcal infections: a controlled trial of 18,738 newborn infants. N Engl J Med 1980;303:769–75.
11. Siegel JD, McCracken GH Jr, Threlkeld N, DePasse BM, Rosenfeld CR. Single dose penicillin prophylaxis of neonatal group B streptococcal disease. Lancet 1982;1:1426–30.
12. Siegel JD, Cushion ND. Prevention of early-onset group B streptococcal disease: another look at single-dose penicillin at birth. Obstet Gynecol 1996;87:692–8.
13. Wedgewood JF, Carlin EB, Benjamin BL, Weintraub A, Lasker M, Green RS, et al.. Penicillin at birth can help prevent early-onset group B streptococcal disease. Pediatrics 1997;99:651–2.
14. Verani JR, McGee L, Schrag SJ; Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC). Prevention of perinatal group B streptococcal disease—revised guidelines from CDC, 2010. MMWR Recomm Rep 2010;59:1–36.
15. American Academy of Pediatrics. Statement of endorsement perinatal group B streptococcal disease. Pediatrics 2011;127:396.
16. Prevention of early-onset group B streptococcal disease in newborns. Committee Opinion No. 485. American College of Obstetricians and Gynecologists. Obstet Gynecol 2011;117:1019–27.
17. Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS, et al.. A population based comparison of strategies to prevent early-onset group B streptococcal disease in neonates. N Engl J Med 2002;347:233–9.
18. Puopolo KM, Madoff LC, Eichenwald EC. Early-onset group B streptococcal screening disease in the era of maternal screening. Pediatrics 2005;115:1240–6.
19. Towers CV, Rumney PJ, Asrat T, Preslicka C, Ghamsary MG, Nageotte MP. The accuracy of late third-trimester antenatal screening for group B streptococcus in predicting colonization at delivery. Am J Perinatol 2010;27:785–90.
20. Kovavisarach E, Jarupisarnlert P, Kanjanaharuetai S. The accuracy of late antenatal screening cultures in predicting intrapartum group B streptococcal colonization. J Med Assoc Thai 2008;91:1796–800.
21. Wendel GD Jr, Leveno KJ, Sánchez PJ, Jackson GL, McIntire DD, Siegel JD. Prevention of neonatal group B streptococcal disease: a combined intrapartum and neonatal protocol. Am J Obstet Gynecol 2002;186:618–26.
22. Velaphi S, Siegel JD, Wendel GD Jr, Cushion N, Eid WM, Sánchez PJ. Early-onset group B streptococcal infection after a combined maternal and neonatal group B streptococcal chemoprophylaxis strategy. Pediatrics 2003;111:541–7.
23. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8.
24. Schuchat A, Zywicki SS, Dinsmoor MJ, Mercer B, Romaquera J, O'Sullivan MJ, et al.. Risk factor and opportunities for prevention of early-onset neonatal sepsis: a multicenter case–control study. Pediatrics 2000;105:21–6.
25. Bromberger P, Lawrence JM, Braun D, Saunders B, Contreras R, Petitti DB. The influence of intrapartum antibiotics on the clinical spectrum of early-onset group B streptococcal infection in term infants. Pediatrics 2000;106:244–50.
26. Puopolo K, Eichenwald E. No change in the incidence of ampicillin-resistant, neonatal early-onset sepsis over 18 years. Pediatrics 2010;125:e1031–8.
27. Sutkin G, Krohn MA, Heine RP, Sweet RL. Antibiotic prophylaxis and non-group B streptococcal neonatal sepsis. Obstet Gynecol 2005;105:581–6.
28. Angstetra D, Ferguson J, Giles WB. Institution of universal screening for group B streptococcus (GBS) from a risk management protocol results in reduction of early-onset GBS disease in a tertiary obstetric unit. Aust N Z J Obstet Gynaecol 2007;47:378–82.